CN110134860B

CN110134860B - User portrait generation method, device and equipment

Info

Publication number: CN110134860B
Application number: CN201910293802.9A
Authority: CN
Inventors: 黄馨誉
Original assignee: Advanced New Technologies Co Ltd
Current assignee: Advanced New Technologies Co Ltd; Advantageous New Technologies Co Ltd
Priority date: 2019-04-12
Filing date: 2019-04-12
Publication date: 2023-04-07
Anticipated expiration: 2039-04-12
Also published as: CN110134860A

Abstract

The embodiment of the application provides a user portrait generation method, a device and equipment, wherein the method comprises the following steps: determining a first service and a second service which is mutually associated with the service content of the first service; acquiring first user behavior data of a user under a first service, second user behavior data of the user under a second service, first service data of the first service, second service data of the second service and external environment data corresponding to the first service; the external environment data is predetermined data related to first service data of a first service; and determining a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data through a pre-trained user portrait determination model. By the embodiment, the user portrait can be accurately determined under the condition that the user relates to a plurality of business fields.

Description

User portrait generation method, device and equipment

Technical Field

The application relates to the technical field of computers, in particular to a user portrait generation method, device and equipment.

Background

A user representation is a descriptive label for a user that is built up in multiple dimensions, and the user representation may represent various attributes of the user. For example, the user representation may include age, gender, whether to buy a house, whether to have children, financial resistance to risk, and the like. By determining the user profile, user requirements can be explored, user preferences can be analyzed, and therefore a more matched internet service can be provided for the user.

With the increase of internet services used by users, the business fields related to the users are gradually increased, and on the basis, a technical scheme is needed to be provided so as to accurately determine the user portrait under the condition that the users relate to a plurality of business fields.

Disclosure of Invention

The embodiment of the application aims to provide a user portrait generation method, device and equipment so as to accurately determine a user portrait under the condition that a user relates to multiple business fields.

In order to achieve the technical purpose, the embodiment of the application is realized as follows:

the embodiment of the application provides a user portrait generation method, which comprises the following steps:

determining a first service and a second service which is mutually associated with the service content of the first service;

acquiring first user behavior data of a user under the first service, second user behavior data of the user under the second service, first service data of the first service, second service data of the second service and external environment data corresponding to the first service; wherein the external environment data is predetermined data related to first service data of the first service;

and determining a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data through a pre-trained user portrait determination model.

The embodiment of the application provides a user portrait generating device, includes:

the service determining module is used for determining a first service and a second service which is related to the service content of the first service;

a data obtaining module, configured to obtain first user behavior data of a user in the first service, second user behavior data of the user in the second service, first service data of the first service, second service data of the second service, and external environment data corresponding to the first service; wherein the external environment data is predetermined data related to first service data of the first service;

and the portrait determining module is used for determining a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data through a pre-trained user portrait determining model.

The embodiment of the application provides a user portrait generating device, which comprises: a processor; and a memory arranged to store computer executable instructions which, when executed, cause the processor to implement the steps of the user representation generation method described above.

Embodiments of the present application provide a storage medium for storing computer-executable instructions, which when executed implement the steps of the user representation generation method described above.

Therefore, through the embodiment, the first service and the second service which is associated with the service content of the first service can be determined, and the target user portrait of the user in the first service is determined according to the first user behavior data of the user in the first service, the second user behavior data of the user in the second service, the first service data of the first service, the second service data of the second service and the external environment data corresponding to the first service through the pre-trained user portrait determination model, so that the user portrait of the user in a certain service field is accurately determined by combining the multi-dimensional data of each service field under the condition that the user relates to a plurality of service fields.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.

FIG. 1 is a schematic flow chart illustrating a user representation generation method according to an embodiment of the present disclosure;

FIG. 2 is a schematic flow chart illustrating a user representation generation method according to another embodiment of the present application;

FIG. 3 is a schematic flow chart illustrating a user representation generation method according to another embodiment of the present application;

FIG. 4 is a block diagram of a user representation generation apparatus according to an embodiment of the present disclosure;

FIG. 5 is a schematic structural diagram of a user portrait determination apparatus according to an embodiment of the present application.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The embodiment of the application provides a user portrait generation method, device and equipment, so that the user portrait can be accurately determined under the condition that a user relates to multiple service fields. The user representation generation method in embodiments of the present application may be performed by a particular device, which may be a computer device for predicting a user representation.

Fig. 1 is a schematic flow chart of a user portrait generation method according to an embodiment of the present application, as shown in fig. 1, the method includes the following steps:

step S102, determining a first service and a second service which is related to the service content of the first service;

step S104, acquiring first user behavior data of a user under a first service, second user behavior data of the user under a second service, first service data of the first service, second service data of the second service and external environment data corresponding to the first service; the external environment data is predetermined data related to first service data of a first service;

and step S106, determining a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data through a pre-trained user portrait determination model.

In the step S102, the specific device determines the first service and the second service associated with the service content of the first service. In this embodiment, the service content of the first service and the service content of the second service are associated with each other, and the association may be embodied as that the service content of the first service and the service content of the second service belong to the same content category, or that both the service content of the first service and the service content of the second service are associated with the same attribute of the user.

For example, the first service is a fund service, and the second service includes, but is not limited to, a stock service, an insurance service, a financial service, a credit card service (e.g. debit), an e-commerce service (e.g. XX shopping mall), and an entertainment service (e.g. XX video website). The business contents of the stock business, the insurance business, the financing business, the credit card business and the fund business belong to financial contents, and the business contents of the E-business, the entertainment business and the fund business are related to the currency attribute of the user.

In step S104, the specific device obtains first user behavior data of the user under the first service, second user behavior data of the user under the second service, first service data of the first service, second service data of the second service, and external environment data corresponding to the first service.

Taking the first service as a fund service and the second service as an example, the second service comprises an e-commerce service and a financial service, the first user behavior data comprises but is not limited to the behavior data of buying fund and redeeming fund, and the second user behavior data comprises but is not limited to the behavior data of buying goods and buying financial behavior data.

The first business data of the first business includes, but is not limited to, product marketing data, discount promotion data, product content data, new data on the product, etc. of the first business, and the second business data of the second business includes, but is not limited to, product marketing data, discount promotion data, product content data, new data on the product, etc. of the second business.

The external environment data corresponding to the first service is predetermined data related to the first service data of the first service, and may specifically be data affecting the first service data. Taking the first service as an example of a fund service, the external environment data corresponding to the first service includes, but is not limited to, interest rate, exchange rate, total Domestic Product GDP (Gross customer Product), consumer Price Index CPI (Consumer Price Index), and the like, and taking the first service as an example of a power business, the external environment data corresponding to the first service includes, but is not limited to, data such as tariff regulation data, power business public opinion data, and Product cost Price fluctuation data.

In step S106, a pre-trained user portrait determination model is used to determine a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data.

In this embodiment, a user portrait determination model is trained in advance. The user representation determination model may be a deep learning based neural network model. The user portrait determining model comprises a first sub model and a second sub model, the first sub model is used for determining an initial user portrait of a user under a first service, the second sub model is used for expanding the initial user portrait based on a second service, an end user portrait of the user under the first service expanded based on the second service is determined, and the end user portrait is a target user portrait. The distinction between the initial user representation and the target user representation will be described in detail below.

The first submodel may be trained by: the method comprises the steps of obtaining first historical behavior data of a user in different periods under a first business, first historical business data of the first business in different periods, historical external environment data of the first business in different periods and initial historical portraits of the user in different periods under the first business as training sample data, inputting the training sample data into a first sub-model for training, and determining that the training of the first sub-model is completed after model parameters of the first sub-model are converged. The process of training the first submodel can be understood as a process of machine learning of the relationship among the first historical behavior data, the first historical service data, the historical external environment data and the initial historical portrait, and the initial user portrait of the user under the first service can be determined based on the behavior data of the user under the first service, the service data of the first service and the external environment data of the first service through the trained first submodel.

When the first submodel is trained, the first historical behavior data in different periods, the first historical business data in different periods, the historical external environment data in different periods and the initial historical portraits in different periods are used as training sample data, so that the trained first submodel can achieve the effect of predicting the initial portraits of users in each time period.

The second submodel may be trained by: the method comprises the steps of obtaining first historical behavior data of a user in different periods under a first business, first historical business data of the first business in different periods, second historical behavior data of the user in different periods under a second business, second historical business data of the second business in different periods, initial historical portraits of the user in different periods under the first business, and target historical portraits of the user in different periods under the first business after the second business is expanded to serve as training sample data, inputting the training sample data into a second submodel for training, and determining that the training of the second submodel is completed after model parameters of the second submodel are converged. The process of training the second submodel may be understood as: and learning a first relation between the first historical behavior data and the second historical behavior data according to the first historical business data and the second historical business data, learning a second relation between the initial historical portrait and the target historical portrait, and fitting the first relation and the second relation. Through the trained second sub-model, the initial user portrait of the user under the first service can be expanded based on the behavior data of the user under the first service, the service data of the first service, the behavior data of the user under the second service and the service data of the second service, and the target user portrait of the user under the first service expanded based on the second service is determined.

When the second submodel is trained, the first historical behavior data in different periods, the first historical business data in different periods, the second historical behavior data in different periods, the second historical business data in different periods, the initial historical portrait in different periods and the target historical portrait in different periods are used as training sample data, so that the trained second submodel can achieve the effect of predicting the portrait of the user in each period.

Based on the above mentioned user portrait determination model, in step S106, determining a target user portrait of the user under the first service according to the first user behavior data, the second user behavior data, the first service data, the second service data and the external environment data, specifically including:

(a1) Determining an initial user portrait of the user under the first service according to the first user behavior data, the first service data and the external environment data;

(a2) And according to the first service data, the second service data, the first user behavior data and the second user behavior data, expanding the initial user portrait to obtain a target user portrait of the user under the first service.

In the above act (a 1), an initial user profile of the user in the first service is determined according to the first user behavior data, the first service data, and the external environment data by using the first sub-model in the user profile determination model. In the above operation (a 2), the initial user portrait is expanded through the second sub-model in the user portrait determination model according to the first service data, the second service data, the first user behavior data and the second user behavior data, so as to obtain the target user portrait of the user under the first service.

The initial user representation is distinguished from the target user representation in that the initial user representation is a user representation determined taking into account only relevant data of the first service, and the target user representation is a representation of the user under the first service determined taking into account the influence of the second service on the user. For example, the first service is a fund service, the second service comprises a financial service and an e-commerce service, the initial user portrait is a portrait of the user under the fund service determined under the condition of only considering relevant data of the fund service, and comprises a user carrying risk capability value, the target user portrait is a portrait of the user under the fund service determined after considering the influence of the financial service and the e-commerce service on the user, and comprises the user carrying risk capability value, a possible probability value of the user for taking out funds for shopping during a short period, a probability value of the user for taking out funds for buying the funds, and the like.

The specific process of the above action (a 1) is as follows:

(a11) Determining a first incidence relation between the first business data and the first user behavior data, and determining a second incidence relation between the external environment data and the first user behavior data;

(a12) And determining an initial user portrait of the user under the first service according to the first association relation and the second association relation.

Specifically, the first association relationship includes, but is not limited to, a positive association and a negative association, the positive association indicates that the expected user behavior corresponding to the first service data matches with the behavior corresponding to the first user behavior data, that is, the user executes the expected behavior or a similar behavior under the influence of the first service data, and the negative association indicates that the expected user behavior corresponding to the first service data is not the behavior corresponding to the first user behavior data, that is, the user does not execute the expected behavior or the similar behavior under the influence of the first service data. The expected user behavior corresponding to the first service data may be set manually.

Similarly, the second association relationship includes, but is not limited to, a positive association and a negative association, the positive association indicates that the expected user behavior corresponding to the external environment data matches with the behavior corresponding to the second user behavior data, that is, the user executes the expected behavior or the like under the influence of the external environment data, and the negative association indicates that the expected user behavior corresponding to the external environment data is not the behavior corresponding to the second user behavior data, that is, the user does not execute the expected behavior or the like under the influence of the external environment data. The desired user behavior corresponding to the external environment data may be set manually.

According to the first incidence relation and the second incidence relation, the influence of the first service data on the user behavior and the influence of the external environment data on the user behavior can be determined, and therefore the initial user portrait of the user under the first service is determined.

In one specific example, the first service is a fund service, the first service data is a fund product release without a new high interest rate, the external environment data is a deposit interest rate increase, and the first user behavior data comprises: the user's behavioral data of redeeming the fund product and the user's behavioral data of browsing the deposit products of the respective banks. Determining, by the act (a 11), that a first association between the first business data and the first user behavior data is a negative association, determining, by the act (a 12), that a second association between the external environment data and the first user behavior data is a positive association, and determining, by the act (a 12), an initial user representation of the user under the fund service based on the first association and the second association comprises: users invest conservatively, tending towards low risk deposits rather than high risk funds. In this example, the expected user behavior corresponding to the first service data includes: and purchasing the existing fund, wherein the expected user behavior corresponding to the external environment data comprises: a deposit is made.

In another specific example, the first service is an e-commerce service, the first service data is the original price sale of e-commerce commodities, the external environment data is the internet-transmitted e-commerce commodities with no discount and counterfeit goods, and the first user behavior data includes behavior data of recharging to an account by a user. Determining, by the act (a 11), that a first correlation between the first business data and the first user behavior data is a negative correlation, determining that a second correlation between the external environment data and the first user behavior data is a negative correlation, and determining, by the act (a 12), an initial user profile of the user under the e-commerce business from the first correlation and the second correlation comprises: e-commerce loyalty users and shopping heavy users. In this example, the expected user behavior corresponding to the first service data includes: the expected user behavior corresponding to the external environment data comprises: not consumed on the e-commerce platform.

As can be seen, through the above actions (a 11) and (a 12), the initial user representation of the user under the first service can be accurately determined based on the first user behavior data, the first service data, and the external environment data.

The specific process of the above-mentioned action (a 2) is:

(a21) Determining a third association relation between the first user behavior data and the second user behavior data according to the first service data and the second service data;

(a22) And expanding the initial user portrait according to the third correlation relation to obtain a target user portrait.

Specifically, the third association is used to indicate whether a causal relationship exists between the first user behavior data and the second user behavior data, and the third association includes: a causal relationship exists between the first user behavior data and the second user behavior data, and no causal relationship exists between the first user behavior data and the second user behavior data. The behavior of the user is influenced by the business data, and whether a causal relationship exists between the first user behavior data and the second user behavior data can be determined by analyzing the first business data and the second business data. Further, the initial user portrait can be expanded according to the third association relation, and a target user portrait expanded based on the second service is obtained.

In one specific example, the first service is a fund service, the first service data is the release of fund products without new high interest rates, the external environment data is the deposit interest rate increase, and the first user behavior data comprises: the act of determining an initial user representation by the above process comprises: users invest conservatively, tending towards low risk deposits rather than high risk funds. Further, the second service is an e-commerce service, and the second user behavior data includes: shopping behavior data, the second business data comprising: and the sales promotion is realized by discounting the commodity by twenty-one. Determining, by the action (a 21), that a causal relationship may exist between the first user behavior data and the second user behavior data based on the first service data and the second service data, and expanding the initial user representation based on the causal relationship to obtain the target user representation in order that the user may take out the fund to shop, the method may include: users invest in less conservative, low risk deposits rather than high risk funds, and prefer shopping over financing.

In another specific example, the first service is an e-commerce service, the first service data is the original price sale of e-commerce commodities, the external environment data is the internet-transmitted e-commerce commodities with no discount and counterfeit goods, the first user behavior data includes behavior data of a user recharging into an account, and determining the initial user profile through the above process includes: e-commerce faithful users and shopping heavy users. Further, the second service is a financial service, and the second user behavior data includes: purchasing financing behavior data, the second business data comprising: no new financing is on the market. Determining that no causal relationship exists between the first user behavior data and the second user behavior data and no direct relationship exists between the user recharging behavior and the user purchasing financing behavior according to the first service data and the second service data through the action (a 21), so as to expand the initial user portrait and obtain the target user portrait, wherein: e-commerce loyalty users, shopping heavy users and available funds are high.

As can be seen, through the above actions (a 21) and (a 22), the initial user image of the user can be expanded based on the second user behavior data and the second service data, so as to obtain the target user image of the user under the first service.

Further, in this embodiment, in consideration of the accuracy of the user portrait determination model, fig. 2 is a schematic flow chart of a user portrait generation method according to another embodiment of the present application, as shown in fig. 2, the method further includes the following steps based on fig. 1:

step S108, after the target user portrait of the user is determined, third user behavior data of the user under the first service is obtained; the third user behavior data and the first user behavior data are different behavior data;

step S110, if the third user behavior data does not match the target user portrait of the user, the target user portrait of the user is corrected according to the third user behavior data, and the user portrait determination model is corrected according to the third user behavior data.

In this embodiment, after the target user portrait of the user is determined, third user behavior data of the user under the first service is also obtained, where the third user behavior data is different from the first user behavior data. The user behavior corresponding to the third user behavior data may occur within a predetermined time period after the target user representation of the user is determined to be complete, e.g., the user behavior corresponding to the third user behavior data may occur within 72 hours after the target user representation of the user is determined to be complete. Alternatively, the user behavior corresponding to the third user behavior data may occur within a predetermined time period before the target user representation of the user is determined to be complete, e.g., the user behavior corresponding to the third user behavior data may occur within 48 hours before the target user representation of the user is determined to be complete. In a preferred embodiment, the third user behavior data may be real-time behavior data of the user.

After the third user behavior data is obtained, whether the third user behavior data is matched with the target user portrait of the user or not is judged, if yes, the target user portrait is accurately predicted, and if not, the target user portrait is wrongly predicted. If the third user behavior data does not match the target user representation, in this embodiment, the target user representation of the user may be corrected based on the third user behavior data, and the user representation-determining model may be corrected based on the third user behavior data.

Correcting the target user representation of the user according to the third user behavior data may be: and deleting the user tags which are not matched with the third user behavior data in the target user image, and adding the user tags corresponding to the third user behavior data to the target user image.

The user profile determination model is calibrated based on the third user behavior data, and may be: and determining a first sample user portrait matching the third user behavior data, and training a user portrait determination model by using the third user behavior data and the first sample user portrait to correct the user portrait determination model.

In this embodiment, the user representation (including the initial user representation, the target user representation, and the first sample user representation) may each include a plurality of user tags, which may be, for example: age, gender, shopping preferences, economic ability, ability to resist risk, etc. In this embodiment, when the user portrait determination model is corrected according to the third user behavior data, one or more user tags matching the third user behavior data may be determined, the determined user tags may be combined into the first user portrait, and the third user behavior data and the first user portrait may be input into the user portrait determination model to train the user portrait determination model, thereby correcting the user portrait determination model.

Therefore, according to the embodiment, after the target user portrait is determined, the target user portrait can be corrected according to the behavior data of the user, and the user portrait determination model can be corrected, so that the effect of feedback adjustment of the user portrait determination model is achieved. If the target user portrait is corrected according to the real-time behavior data of the user and the user portrait determination model is corrected, the user portrait determination model can achieve the function of continuously predicting the user portrait within 7 x 24 hours.

Further, in this embodiment, in consideration of the accuracy of the user portrait determination model, fig. 3 is a schematic flow chart of a user portrait generation method according to another embodiment of the present application, as shown in fig. 3, the method further includes the following steps based on fig. 1:

step S112, after the target user image of the user is determined, the same type of user belonging to the same preset type with the user under the first service is determined, and the target user image of the same type of user is obtained;

step S114, if the difference between the target user portrait of the user and the target user portrait of the same type of user is larger than the preset difference, the user portrait determination model is corrected according to the user behavior data of the same type of user.

In this embodiment, after determining the target user profile of the user, a similar user belonging to the same preset category as the user under the first service is also determined, where the similar user may be, for example, a user belonging to a severe purchasing fan as the user under the e-commerce service, or may be, for example, a user belonging to a low risk resistance as the user under the fund service. And then, acquiring the target user image of the same type of user. And comparing whether the difference between the target user portrait of the user and the target user portrait of the same type of user is greater than a preset difference, specifically, comparing whether each user label included in the target user portrait of the user is overlapped with each user label included in the target user portrait of the same type of user, determining the overlapping rate of the user labels, if the overlapping rate is greater than a preset overlapping rate threshold value, determining that the difference is less than the preset difference, otherwise, determining that the difference is greater than the preset difference. If the difference between the target user portrait of the user and the target user portrait of the similar user is larger than the preset difference, it is indicated that the user portrait determination model is not accurate enough, and then user behavior data of the similar user is obtained, and the user portrait determination model is corrected according to the user behavior data of the similar user.

The user behavior corresponding to the user behavior data of the same type of user may occur within a preset time period after the target user profile of the same type of user is determined, for example, the user behavior corresponding to the user behavior data of the same type of user may occur within 72 hours after the target user profile of the same type of user is determined. Alternatively, the user behavior corresponding to the user behavior data of the similar user may occur within a preset time period before the target user image determination of the similar user is completed, for example, the user behavior corresponding to the user behavior data of the similar user occurs within 48 hours before the target user image determination of the similar user is completed. In a preferred embodiment, the user behavior data of the same type of users may be real-time behavior data of the same type of users.

According to the user behavior data of the same type of users, the user portrait determination model is corrected, and the method specifically comprises the following steps: a second sample user representation matching the user behavior data of the same type of user is determined, and the user representation determination model is trained using the user behavior data of the same type of user and the second sample user representation to correct the user representation determination model.

In this embodiment, the user representation (including the initial user representation, the target user representation, and the second sample user representation) may each include a plurality of user tags, which may be, for example: age, gender, shopping preferences, economic ability, resistance to risks, etc. In this embodiment, when the user portrait determination model is corrected according to the user behavior data of the same type of user, one or more user tags matched with the user behavior data of the same type of user may be determined, the determined user tags may be combined into a second sample user portrait, and the user behavior data of the same type of user and the second sample user portrait are input into the user portrait determination model to train the user portrait determination model, thereby correcting the user portrait determination model.

Therefore, by the embodiment, after the target user portrait of the user is determined, the user portrait determination model can be corrected according to the user behavior data of the same type of user, and the effect of transversely comparing, feeding back and adjusting the user portrait determination model is achieved.

In summary, the method in the embodiment of the present application has at least the following beneficial effects: the service data of a plurality of services and the behavior data of the user under the services are accessed, and the services are opened, so that the user portrait description is more comprehensive; external environment data are introduced, so that a feature space can be enlarged, and the portrait description is more accurate; by transversely comparing the user behavior data with the similar users, the model can be corrected in time, and the effect of depicting the user portrait within 7 × 24 hours is achieved.

Corresponding to the user portrait generation method, an embodiment of the present application further provides a user portrait generation apparatus for implementing the user portrait generation method, fig. 4 is a schematic diagram of module components of the user portrait generation apparatus provided in an embodiment of the present application, as shown in fig. 4, the apparatus includes the following modules:

a service determination module 41, configured to determine a first service and a second service that is associated with service content of the first service;

a data obtaining module 42, configured to obtain first user behavior data of a user in the first service, second user behavior data of the user in the second service, first service data of the first service, second service data of the second service, and external environment data corresponding to the first service; wherein the external environment data is predetermined data related to first service data of the first service;

a representation determining module 43, configured to determine, according to the first user behavior data, the second user behavior data, the first service data, the second service data, and the external environment data, a target user representation of the user under the first service through a pre-trained user representation determining model.

Optionally, the portrait determination module 43 is specifically configured to: determining an initial user representation of the user under the first service according to the first user behavior data, the first service data and the external environment data; and expanding the initial user portrait according to the first service data, the second service data, the first user behavior data and the second user behavior data to obtain a target user portrait of the user under the first service.

Optionally, the portrait determination module 43 is further specifically configured to: determining a first association relationship between the first business data and the first user behavior data, and determining a second association relationship between the external environment data and the first user behavior data; and determining an initial user portrait of the user under the first service according to the first association relation and the second association relation.

Optionally, the portrait determination module 43 is further specifically configured to: determining a third association relationship between the first user behavior data and the second user behavior data according to the first service data and the second service data; and expanding the initial user portrait according to the third association relation to obtain the target user portrait.

Optionally, the apparatus further comprises a first correction module configured to: after determining the target user portrait of the user, acquiring third user behavior data of the user under the first service; wherein the third user behavior data is different behavior data from the first user behavior data; if the third user behavior data does not match the target user representation of the user, then the target user representation of the user is corrected according to the third user behavior data, and the user representation-determining model is corrected according to the third user behavior data.

Optionally, the first correction module is specifically configured to: determining a first sample user representation that matches the third user behavior data; training the user representation determination model using the third user behavior data and the first sample user representation to correct the user representation determination model.

Optionally, the apparatus further comprises a second correction module, configured to: after determining the target user image of the user, determining the same type of user belonging to the same preset category as the user under the first service, and acquiring the target user image of the same type of user; and if the difference between the target user portrait of the user and the target user portrait of the same type of user is larger than a preset difference, correcting the user portrait determination model according to the user behavior data of the same type of user.

Optionally, the second correction module is specifically configured to: determining a second sample user representation that matches the user behavior data of the homogeneous user; training the user representation determination model using the user behavior data of the homogeneous user and the second sample user representation to correct the user representation determination model.

Therefore, according to the embodiment, the first service and the second service which is mutually associated with the service content of the first service can be determined, and the target user portrait of the user under the first service is determined according to the first user behavior data of the user under the first service, the second user behavior data of the user under the second service, the first service data of the first service, the second service data of the second service and the external environment data corresponding to the first service through the pre-trained user portrait determination model, so that the user portrait of the user under a certain service field is accurately determined by combining multi-dimensional data of each service field under the condition that the user relates to multiple service fields.

It should be noted that the user portrait determining apparatus in the embodiment of the present application can implement each process of the foregoing user portrait determining method embodiment, and achieve the same function and effect, which is not described herein again.

Further, an embodiment of the present application also provides a user portrait determination apparatus, fig. 5 is a schematic structural diagram of the user portrait determination apparatus provided in an embodiment of the present application, and as shown in fig. 5, the user portrait determination apparatus may generate a relatively large difference due to different configurations or performances, and may include one or more processors 901 and a memory 902, where the memory 902 may store one or more stored applications or data. Memory 902 may be, among other things, transient storage or persistent storage. The application program stored in memory 902 may include one or more modules (not shown), each of which may include a series of computer-executable instructions for a user representation determination device. Still further, processor 901 may be configured to communicate with memory 902 to execute a series of computer-executable instructions in memory 902 on a user representation determining device. The user representation determination apparatus may also include one or more power supplies 903, one or more wired or wireless network interfaces 904, one or more input-output interfaces 905, one or more keyboards 906, and the like.

In one particular embodiment, a user representation determination apparatus includes memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the user representation determination apparatus, and execution of the one or more programs by one or more processors includes computer-executable instructions for:

acquiring first user behavior data of a user under the first service, second user behavior data of the user under the second service, first service data of the first service, second service data of the second service and external environment data corresponding to the first service; the external environment data is predetermined data related to first service data of the first service;

Optionally, when executed, the computer executable instructions determine a target user representation of the user under the first service based on the first user behavior data, the second user behavior data, the first service data, the second service data, and the external environment data, including: determining an initial user portrait of the user under the first service according to the first user behavior data, the first service data and the external environment data; and according to the first service data, the second service data, the first user behavior data and the second user behavior data, expanding the initial user portrait to obtain a target user portrait of the user under the first service.

Optionally, computer-executable instructions, when executed, determine an initial user representation of the user under the first service from the first user behavior data, the first service data, and the external environment data, comprising: determining a first incidence relation between the first business data and the first user behavior data, and determining a second incidence relation between the external environment data and the first user behavior data; and determining an initial user portrait of the user under the first service according to the first association relation and the second association relation.

Optionally, when executed, the computer-executable instructions expand the initial user representation according to the first service data, the second service data, the first user behavior data, and the second user behavior data, to obtain a target user representation of the user under the first service, including: determining a third association relationship between the first user behavior data and the second user behavior data according to the first service data and the second service data; and expanding the initial user portrait according to the third association relation to obtain the target user portrait.

Optionally, the computer executable instructions, when executed, further comprise: after determining a target user portrait of the user, acquiring third user behavior data of the user under the first service; wherein the third user behavior data is different behavior data from the first user behavior data; if the third user behavior data does not match the target user representation of the user, then the target user representation of the user is corrected according to the third user behavior data, and the user representation-determining model is corrected according to the third user behavior data.

Optionally, computer executable instructions, when executed, correct the user representation determination model based on the third user behavior data, comprising: determining a first sample user representation that matches the third user behavior data; training the user representation determination model using the third user behavior data and the first sample user representation to correct the user representation determination model.

Optionally, the computer executable instructions, when executed, further comprise: after the target user image of the user is determined, determining similar users belonging to the same preset category as the user under the first service, and acquiring the target user image of the similar users; and if the difference between the target user portrait of the user and the target user portrait of the same type of user is larger than a preset difference, correcting the user portrait determination model according to the user behavior data of the same type of user.

Optionally, the computer executable instructions, when executed, correct the user representation determination model based on user behavior data of the homogeneous user, comprising: determining a second sample user representation that matches the user behavior data of the homogeneous user; training the user representation determination model using the user behavior data of the homogeneous user and the second sample user representation to correct the user representation determination model.

It should be noted that the user portrait determination device in the embodiment of the present application can implement each process of the foregoing user portrait determination method embodiment, and achieve the same function and effect, which is not described herein again.

Further, an embodiment of the present application further provides a storage medium for storing computer-executable instructions, in a specific embodiment, the storage medium may be a usb disk, an optical disk, a hard disk, and the like, and when executed by a processor, the storage medium stores the computer-executable instructions and can implement the following processes:

determining a first service and a second service which is related to the service content of the first service;

Optionally, the storage medium stores computer-executable instructions that, when executed by a processor, determine a target user representation of the user under the first service based on the first user behavior data, the second user behavior data, the first service data, the second service data, and the external environment data, comprising: determining an initial user representation of the user under the first service according to the first user behavior data, the first service data and the external environment data; and according to the first service data, the second service data, the first user behavior data and the second user behavior data, expanding the initial user portrait to obtain a target user portrait of the user under the first service.

Optionally, the storage medium stores computer-executable instructions that, when executed by a processor, determine an initial user representation of the user under the first service based on the first user behavior data, the first service data, and the external environment data, comprising: determining a first incidence relation between the first business data and the first user behavior data, and determining a second incidence relation between the external environment data and the first user behavior data; and determining an initial user portrait of the user under the first service according to the first association relation and the second association relation.

Optionally, when executed by a processor, the computer-executable instructions stored in the storage medium expand the initial user representation according to the first service data, the second service data, the first user behavior data, and the second user behavior data to obtain a target user representation of the user under the first service, including: determining a third association relationship between the first user behavior data and the second user behavior data according to the first service data and the second service data; and expanding the initial user portrait according to the third association relation to obtain the target user portrait.

Optionally, the storage medium stores computer executable instructions that, when executed by the processor, further comprise: after determining the target user portrait of the user, acquiring third user behavior data of the user under the first service; wherein the third user behavior data is different behavior data from the first user behavior data; and if the third user behavior data does not match the target user representation of the user, correcting the target user representation of the user according to the third user behavior data, and correcting the user representation determination model according to the third user behavior data.

Optionally, the storage medium stores computer-executable instructions that, when executed by the processor, correct the user representation-determining model based on the third user behavior data, comprising: determining a first sample user representation that matches the third user behavior data; training the user representation determination model using the third user behavior data and the first sample user representation to correct the user representation determination model.

Optionally, the storage medium stores computer executable instructions that, when executed by the processor, further comprise: after the target user image of the user is determined, determining similar users belonging to the same preset category as the user under the first service, and acquiring the target user image of the similar users; and if the difference between the target user portrait of the user and the target user portrait of the same type of user is larger than a preset difference, correcting the user portrait determination model according to the user behavior data of the same type of user.

Optionally, the storage medium stores computer-executable instructions that, when executed by the processor, correct the user representation determination model based on user behavior data of the homogeneous user, comprising: determining a second sample user representation that matches the user behavior data of the homogeneous user; training the user representation determination model using the user behavior data of the homogeneous user and the second sample user representation to correct the user representation determination model.

It should be noted that the storage medium in the embodiment of the present application can implement each process of the foregoing user portrait determination method embodiment, and achieve the same function and effect, which is not described herein again.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical blocks. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as ABEL (Advanced Boolean Expression Language), AHDL (alternate Hardware Description Language), traffic, CUPL (core universal Programming Language), HDCal, jhddl (Java Hardware Description Language), lava, lola, HDL, PALASM, rhyd (Hardware Description Language), and vhigh-Language (Hardware Description Language), which is currently used in most popular applications. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A user representation generation method, comprising:

2. The method of claim 1, determining a target user representation of the user under the first service based on the first user behavior data, the second user behavior data, the first service data, the second service data, and the external environment data, comprising:

determining an initial user representation of the user under the first service according to the first user behavior data, the first service data and the external environment data;

and according to the first service data, the second service data, the first user behavior data and the second user behavior data, expanding the initial user portrait to obtain a target user portrait of the user under the first service.

3. The method of claim 2, determining an initial user representation of the user under the first service from the first user behavior data, the first service data, and the external environment data, comprising:

determining a first incidence relation between the first business data and the first user behavior data, and determining a second incidence relation between the external environment data and the first user behavior data;

and determining an initial user portrait of the user under the first service according to the first association relation and the second association relation.

4. The method of claim 2, extending the initial user representation according to the first service data, the second service data, the first user behavior data, and the second user behavior data to obtain a target user representation of the user under the first service, comprising:

determining a third association relationship between the first user behavior data and the second user behavior data according to the first service data and the second service data;

and expanding the initial user portrait according to the third correlation relation to obtain the target user portrait.

5. The method of any of claims 1 to 4, further comprising:

after determining a target user portrait of the user, acquiring third user behavior data of the user under the first service; wherein the third user behavior data is different behavior data from the first user behavior data;

if the third user behavior data does not match the target user representation of the user, then the target user representation of the user is corrected according to the third user behavior data, and the user representation-determining model is corrected according to the third user behavior data.

6. The method of claim 5, correcting the user portrait determination model from the third user behavior data, comprising:

determining a first sample user representation that matches the third user behavior data;

training the user representation determination model using the third user behavior data and the first sample user representation to correct the user representation determination model.

7. The method of any of claims 1 to 4, further comprising:

after determining the target user image of the user, determining the same type of user belonging to the same preset category as the user under the first service, and acquiring the target user image of the same type of user;

and if the difference between the target user portrait of the user and the target user portrait of the same type of user is larger than a preset difference, correcting the user portrait determination model according to the user behavior data of the same type of user.

8. The method of claim 7, correcting the user representation determination model based on user behavior data for the homogeneous user, comprising:

determining a second sample user representation that matches the user behavior data of the homogeneous user;

training the user representation determination model with the user behavior data of the homogeneous user and the second sample user representation to correct the user representation determination model.

9. A user representation generation apparatus, comprising:

10. The apparatus of claim 9, the representation determination module to:

and expanding the initial user portrait according to the first service data, the second service data, the first user behavior data and the second user behavior data to obtain a target user portrait of the user under the first service.

11. The apparatus of claim 10, the representation determination module further to:

12. The apparatus of claim 10, the representation determination module further to:

13. The apparatus of any of claims 9 to 12, further comprising a first correction module to:

after determining the target user portrait of the user, acquiring third user behavior data of the user under the first service; wherein the third user behavior data is different behavior data from the first user behavior data;

14. The apparatus of claim 13, the first correction module to be specifically configured to:

15. The apparatus of any of claims 9 to 12, further comprising a second correction module to:

16. The apparatus of claim 15, the second correction module to be specifically configured to:

training the user representation determination model using the user behavior data of the homogeneous user and the second sample user representation to correct the user representation determination model.

17. A user representation generation apparatus, comprising: a processor; and a memory arranged to store computer executable instructions which, when executed, cause the processor to carry out the steps of the user representation generation method of any preceding claim 1 to 8.

18. A storage medium storing computer-executable instructions which, when executed, implement the steps of the user representation generation method of any of claims 1 to 8.